- Title
- Web of students : class-level friendship network discovery from educational big data
- Creator
- Guo, Teng; Tang, Tang; Zhang, Dongyu; Li, Jianxin; Xia, Feng
- Date
- 2021
- Type
- Text; Conference paper
- Identifier
- http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/183585
- Identifier
- vital:16317
- Identifier
-
https://doi.org/10.1007/978-3-030-90888-1_38
- Identifier
- ISBN:03029743 (ISSN); 9783030908874 (ISBN)
- Abstract
- Classmate friendships are a major aspect of university social experience. Taking classes together is one of the main ways for students to build friendships. Consequently, class-level friendship networks have attracted tremendous attention from researchers. They are also very useful in student support and early intervention. However, these networks are normally invisible for educators. Discovering such an important web of students effectively is a pressing problem. Against this background, we propose a data-driven framework called CANDY which automatically discovers the class-level friendship networks based on educational big data. We first represent features through representation learning methods. Secondly, the data is augmented with the randomly shuffling method. Thirdly, a conditional generative adversarial network model is used to mine the class-level friendship networks. A deep adversarial optimization strategy is proposed here for problems caused by network sparsity. To evaluate the performance of the proposed approach, we build a real-world dataset that contains rich student information. Extensive experiments have been conducted and the results demonstrate the effectiveness of our framework. © 2021, Springer Nature Switzerland AG.
- Publisher
- Springer Science and Business Media Deutschland GmbH
- Relation
- 22nd International Conference on Web Information Systems Engineering, WISE 2021 p. 497-511
- Rights
- All metadata describing materials held in, or linked to, the repository is freely available under a CC0 licence
- Rights
- Copyright © 2021, Springer Nature Switzerland AG.
- Subject
- Educational big data; Friendship networks; Generative adversarial networks; Social network analysis
- Reviewed
- Hits: 410
- Visitors: 391
- Downloads: 0
| Thumbnail | File | Description | Size | Format |
|---|